CHANCE: liver transplantation in high-grade ACLF — where do the limits lie?

Liver transplantation in high-grade acute-on-chronic liver failure (ACLF) presents one of the most difficult decisions in liver critical care. Patients may have profound multiorgan dysfunction, require several forms of life-sustaining support and face an extremely high risk of death without transplantation. Yet transplantation itself represents a major physiological insult, and the likelihood of meaningful recovery must be weighed against the scarcity of available organs.

Previous studies have demonstrated encouraging post-transplant outcomes in patients with ACLF-3. However, much of the supporting evidence has been retrospective, drawn from transplantation registries or limited to individual centres and geographical regions. Differences in the classification of organ failure and the selection of patients for transplantation have made interpretation challenging.

The CHANCE study represents a major contribution to this evidence base. Through an ambitious international collaboration, the investigators prospectively collected detailed clinical data across 63 transplant centres, examining outcomes in both transplanted patients and those assessed but declined for transplantation.

The central questions are how best to select and prioritise patients with high-grade ACLF, and whether the number and type of organ failures can help identify those at greatest postoperative risk.

• Design: Prospective, multicentre, international observational cohort study.
• Population: 995 adults with decompensated cirrhosis enrolled between 2021 and 2023 across 63 liver transplant centres.
• Transplant recipients: 612 patients, including 143 with ACLF-2 and 130 with ACLF-3 at transplantation.
• Comparator: 144 patients with ACLF-2 or ACLF-3 who had been assessed but declined for transplantation.
• Principal outcomes: One-year death without retransplantation after liver transplantation, and one-year mortality after inclusion among patients declined for transplantation.
• Key finding: Post-transplant mortality was 10.5% in ACLF-2 and 20.8% in ACLF-3, increasing to 41.7% with five or six organ failures.

The investigators recruited patients from specialist liver transplant centres in Europe, North America, Latin America and Asia.

Patients were enrolled into three groups: those listed for transplantation with ACLF-2 or ACLF-3; those listed without ACLF or with ACLF-1 and a MELD score greater than 20; and patients with ACLF-2 or ACLF-3 who were assessed but declined for transplantation.

Organ failures were defined using EASL-CLIF criteria, covering liver, kidney, coagulation, cerebral, circulatory and respiratory failure. Patients were assessed prospectively during their admission, and organ-failure status was documented again immediately before transplantation.

This latter detail is particularly important. ACLF is a dynamic syndrome, and the clinical state at listing may differ substantially from that at transplantation.

The principal analyses examined one-year mortality according to ACLF grade and the number of failing organs. Competing-risk analyses accounted for retransplantation, while multivariable models explored the associations between clinical characteristics and postoperative death. Additional analyses from study inclusion included propensity-score matching between listed and non-listed patients.

Patients with high-grade ACLF faced substantial waiting-list risk

Patients listed with ACLF-2 or ACLF-3 were more likely to die while waiting or be delisted because of deterioration than those listed without ACLF or with ACLF-1: 120/437 (27.5%) versus 56/387 (14.5%). The investigators reported 27.4% for the former, a small rounding discrepancy.

This occurred despite shorter waiting times among those ultimately transplanted. Median time from listing to transplantation was seven days for patients with ACLF-3, compared with 30 days for those without ACLF.

ACLF was dynamic. Among transplanted recipients initially listed with ACLF-2 or ACLF-3, 70/301 (23.3%) improved to a lower grade before surgery. Conversely, 43/311 (13.8%) listed without high-grade ACLF deteriorated to a higher grade. The paper reported 13.5% for the latter, although 43/311 rounds to 13.8%.

These findings reinforce the importance of timely transplant access and repeated clinical reassessment.

One-year survival after transplantation was encouraging

Of the 612 patients transplanted, 73 (11.9%) died without retransplantation during the first year, and 13 (2.1%) underwent retransplantation.

These results are striking given the intensity of organ support required by patients with ACLF-3. At transplantation, 63.8% were receiving vasopressors, 65.4% renal replacement therapy and 47.3% mechanical ventilation. These supports frequently overlapped.

Among patients assessed but declined for transplantation, mortality within one year was 73.0% (46/63) in ACLF-2 and 87.7% (71/81) in ACLF-3.

Although the transplanted and non-transplanted populations were not directly comparable, the findings reinforce the potential for transplantation to achieve meaningful survival in appropriately selected patients. Analyses from the time of inclusion, including propensity-score matching, further supported an association between listing and improved survival.

The number of failing organs mattered — particularly beyond four

Among 130 patients transplanted with ACLF-3, 67 had three failing organs, 39 had four and 24 had five or six.

Patients with three or four failures had relatively similar outcomes, whereas those with five or six experienced considerably higher mortality. The separation was apparent early after transplantation and persisted through the first year.

In adjusted analysis, five or more organ failures were associated with higher post-transplant mortality compared with no organ failure (subdistribution hazard ratio 9.46; 95% CI 3.29–27.15).

The highest-risk group was small, comprising just 24 recipients. This is an important prognostic signal, not a validated universal threshold of transplant futility.

The type of organ failure may be important

In unadjusted analyses, one-year mortality was higher among recipients with cerebral failure (28.4%), respiratory failure (25.9%) and circulatory failure (20.5%) than among those with kidney failure (14.2%) or liver failure (13.8%).

These groups overlapped, so these percentages should not be interpreted as isolated effects of individual organ failures.

Cerebral failure remained associated with higher mortality in a multivariable model (subdistribution hazard ratio 2.22; 95% CI 1.04–4.76). Adjusted associations for respiratory and circulatory failure were less certain.

This suggests that type, severity and reversibility may be important alongside the total number of failing organs.

Recovery following transplantation was encouraging

Recipients with ACLF-3 required longer postoperative intensive care stays: a median of 10.5 days versus five days in patients without ACLF.

Bleeding, acute kidney injury and infection were important postoperative complications, and refractory septic shock was the leading recorded cause of post-transplant death.

Nevertheless, among 94 surviving ACLF-3 recipients with available one-year assessment, 84 (89.4%) were outpatients and only two (2.1%) remained dependent on long-term renal replacement therapy.

These observations suggest meaningful recovery is possible after substantial multiorgan dysfunction, although they do not establish complete functional recovery or normal quality of life.

• Can repeated assessment of organ failure identify the optimal transplantation window more reliably than a single ACLF grade?
• Should five or six organ failures be classified separately from three or four, and would this improve prognostication beyond existing scores?
• How should cerebral, respiratory and circulatory dysfunction individually influence transplant selection?
• Can further prospective studies better capture longer-term functional recovery and quality of life?